Data Visualization, Data Reduction and Classifier Fusion for Intelligent Fault Diagnosis in Gas Turbine Engines

In this paper, we investigate four key issues associated with data-driven approaches for fault classification using the Pratt and Whitney commercial dual-spool turbofan engine data as a test case. The four issues considered here include the following. (1) Can we characterize, a priori, the difficulty of fault classification via self-organizing maps? (2) Do data reduction techniques improve fault classification performance and enable the implementation of data-driven classification techniques in memory-constrained digital electronic control units (DECUs)? (3) When does adaptive boosting, an incremental fusion method that successively combines moderately inaccurate classifiers into accurate ones, help improve classification performance? (4) How to synthesize classifier fusion architectures to improve the overall diagnostic accuracy? The classifiers studied in this paper are the support vector machine, probabilistic neural network, $k$-nearest neighbor, principal component analysis, Gaussian mixture models, and a physics-based single fault isolator. As these algorithms operate on large volumes of data and are generally computationally expensive, we reduce the data set using the multiway partial least squares method. This has the added benefits of improved diagnostic accuracy and smaller memory requirements. The performance of the moderately inaccurate classifiers is improved using adaptive boosting. These results are compared to the results of the classifiers alone, as well as different fusion architectures. We show that fusion reduces the variability in diagnostic accuracy, and is most useful when combining moderately inaccurate classifiers.